1 / 119

Lexical Semantics & Word Sense Disambiguation

Lexical Semantics & Word Sense Disambiguation. Ling571 Deep Processing Techniques for NLP February 16, 2011. Roadmap. Lexical semantics Lexical taxonomy WordNet Thematic Roles Issues Resources: PropBank & FrameNet Selectional Restrictions Primitive decompositions. WordNet Taxonomy.

hafwen
Download Presentation

Lexical Semantics & Word Sense Disambiguation

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Lexical Semantics & Word Sense Disambiguation Ling571 Deep Processing Techniques for NLP February 16, 2011

  2. Roadmap • Lexical semantics • Lexical taxonomy • WordNet • Thematic Roles • Issues • Resources: • PropBank& FrameNet • Selectional Restrictions • Primitive decompositions

  3. WordNet Taxonomy • Most widely used English sense resource • Manually constructed lexical database • 3 Tree-structured hierarchies • Nouns (117K) , verbs (11K), adjective+adverb (27K) • Entries: synonym set, gloss, example use • Relations between entries: • Synonymy: in synset • Hypo(per)nym: Isa tree

  4. WordNet

  5. Noun WordNet Relations

  6. WordNet Taxonomy

  7. Thematic Roles • Describe semantic roles of verbal arguments • Capture commonality across verbs

  8. Thematic Roles • Describe semantic roles of verbal arguments • Capture commonality across verbs • E.g. subject of break, open is AGENT • AGENT: volitional cause • THEME: things affected by action

  9. Thematic Roles • Describe semantic roles of verbal arguments • Capture commonality across verbs • E.g. subject of break, open is AGENT • AGENT: volitional cause • THEME: things affected by action • Enables generalization over surface order of arguments • JohnAGENT broke the windowTHEME

  10. Thematic Roles • Describe semantic roles of verbal arguments • Capture commonality across verbs • E.g. subject of break, open is AGENT • AGENT: volitional cause • THEME: things affected by action • Enables generalization over surface order of arguments • JohnAGENT broke the windowTHEME • The rockINSTRUMENTbroke the windowTHEME

  11. Thematic Roles • Describe semantic roles of verbal arguments • Capture commonality across verbs • E.g. subject of break, open is AGENT • AGENT: volitional cause • THEME: things affected by action • Enables generalization over surface order of arguments • JohnAGENT broke the windowTHEME • The rockINSTRUMENTbroke the windowTHEME • The windowTHEME was broken by JohnAGENT

  12. Thematic Roles • Thematic grid, θ-grid, case frame • Set of thematic role arguments of verb

  13. Thematic Roles • Thematic grid, θ-grid, case frame • Set of thematic role arguments of verb • E.g. Subject:AGENT; Object:THEME, or • Subject: INSTR; Object: THEME

  14. Thematic Roles • Thematic grid, θ-grid, case frame • Set of thematic role arguments of verb • E.g. Subject:AGENT; Object:THEME, or • Subject: INSTR; Object: THEME • Verb/Diathesis Alternations • Verbs allow different surface realizations of roles

  15. Thematic Roles • Thematic grid, θ-grid, case frame • Set of thematic role arguments of verb • E.g. Subject:AGENT; Object:THEME, or • Subject: INSTR; Object: THEME • Verb/Diathesis Alternations • Verbs allow different surface realizations of roles • DorisAGENTgave the bookTHEME to CaryGOAL

  16. Thematic Roles • Thematic grid, θ-grid, case frame • Set of thematic role arguments of verb • E.g. Subject:AGENT; Object:THEME, or • Subject: INSTR; Object: THEME • Verb/Diathesis Alternations • Verbs allow different surface realizations of roles • DorisAGENTgave the bookTHEME to CaryGOAL • DorisAGENT gave CaryGOAL the bookTHEME

  17. Thematic Roles • Thematic grid, θ-grid, case frame • Set of thematic role arguments of verb • E.g. Subject:AGENT; Object:THEME, or • Subject: INSTR; Object: THEME • Verb/Diathesis Alternations • Verbs allow different surface realizations of roles • DorisAGENTgave the bookTHEME to CaryGOAL • DorisAGENT gave CaryGOAL the bookTHEME • Group verbs into classes based on shared patterns

  18. Canonical Roles

  19. Thematic Role Issues • Hard to produce

  20. Thematic Role Issues • Hard to produce • Standard set of roles • Fragmentation: Often need to make more specific • E,g, INSTRUMENTS can be subject or not

  21. Thematic Role Issues • Hard to produce • Standard set of roles • Fragmentation: Often need to make more specific • E,g, INSTRUMENTS can be subject or not • Standard definition of roles • Most AGENTs: animate, volitional, sentient, causal • But not all…. • Strategies: • Generalized semantic roles: PROTO-AGENT/PROTO-PATIENT • Defined heuristically: PropBank • Define roles specific to verbs/nouns: FrameNet

  22. Thematic Role Issues • Hard to produce • Standard set of roles • Fragmentation: Often need to make more specific • E,g, INSTRUMENTS can be subject or not • Standard definition of roles • Most AGENTs: animate, volitional, sentient, causal • But not all….

  23. Thematic Role Issues • Hard to produce • Standard set of roles • Fragmentation: Often need to make more specific • E,g, INSTRUMENTS can be subject or not • Standard definition of roles • Most AGENTs: animate, volitional, sentient, causal • But not all…. • Strategies: • Generalized semantic roles: PROTO-AGENT/PROTO-PATIENT • Defined heuristically: PropBank

  24. Thematic Role Issues • Hard to produce • Standard set of roles • Fragmentation: Often need to make more specific • E,g, INSTRUMENTS can be subject or not • Standard definition of roles • Most AGENTs: animate, volitional, sentient, causal • But not all…. • Strategies: • Generalized semantic roles: PROTO-AGENT/PROTO-PATIENT • Defined heuristically: PropBank • Define roles specific to verbs/nouns: FrameNet

  25. PropBank • Sentences annotated with semantic roles • Penn and Chinese Treebank

  26. PropBank • Sentences annotated with semantic roles • Penn and Chinese Treebank • Roles specific to verb sense • Numbered: Arg0, Arg1, Arg2,… • Arg0: PROTO-AGENT; Arg1: PROTO-PATIENT, etc

  27. PropBank • Sentences annotated with semantic roles • Penn and Chinese Treebank • Roles specific to verb sense • Numbered: Arg0, Arg1, Arg2,… • Arg0: PROTO-AGENT; Arg1: PROTO-PATIENT, etc • E.g. agree.01 • Arg0: Agreer

  28. PropBank • Sentences annotated with semantic roles • Penn and Chinese Treebank • Roles specific to verb sense • Numbered: Arg0, Arg1, Arg2,… • Arg0: PROTO-AGENT; Arg1: PROTO-PATIENT, etc • E.g. agree.01 • Arg0: Agreer • Arg1: Proposition

  29. PropBank • Sentences annotated with semantic roles • Penn and Chinese Treebank • Roles specific to verb sense • Numbered: Arg0, Arg1, Arg2,… • Arg0: PROTO-AGENT; Arg1: PROTO-PATIENT, etc • E.g. agree.01 • Arg0: Agreer • Arg1: Proposition • Arg2: Other entity agreeing

  30. PropBank • Sentences annotated with semantic roles • Penn and Chinese Treebank • Roles specific to verb sense • Numbered: Arg0, Arg1, Arg2,… • Arg0: PROTO-AGENT; Arg1: PROTO-PATIENT, etc • E.g. agree.01 • Arg0: Agreer • Arg1: Proposition • Arg2: Other entity agreeing • Ex1: [Arg0The group] agreed [Arg1it wouldn’t make an offer]

  31. FrameNet • Semantic roles specific to Frame • Frame: script-like structure, roles (frame elements)

  32. FrameNet • Semantic roles specific to Frame • Frame: script-like structure, roles (frame elements) • E.g. change_position_on_scale: increase, rise • Attribute, Initial_value, Final_value

  33. FrameNet • Semantic roles specific to Frame • Frame: script-like structure, roles (frame elements) • E.g. change_position_on_scale: increase, rise • Attribute, Initial_value, Final_value • Core, non-core roles

  34. FrameNet • Semantic roles specific to Frame • Frame: script-like structure, roles (frame elements) • E.g. change_position_on_scale: increase, rise • Attribute, Initial_value, Final_value • Core, non-core roles • Relationships b/t frames, frame elements • Add causative: cause_change_position_on_scale

  35. Selectional Restrictions • Semantic type constraint on arguments • I want to eat someplace close to UW

  36. Selectional Restrictions • Semantic type constraint on arguments • I want to eat someplace close to UW • E.g. THEME of eating should be edible • Associated with senses

  37. Selectional Restrictions • Semantic type constraint on arguments • I want to eat someplace close to UW • E.g. THEME of eating should be edible • Associated with senses • Vary in specificity:

  38. Selectional Restrictions • Semantic type constraint on arguments • I want to eat someplace close to UW • E.g. THEME of eating should be edible • Associated with senses • Vary in specificity: • Imagine: AGENT: human/sentient; THEME: any

  39. Selectional Restrictions • Semantic type constraint on arguments • I want to eat someplace close to UW • E.g. THEME of eating should be edible • Associated with senses • Vary in specificity: • Imagine: AGENT: human/sentient; THEME: any • Representation: • Add as predicate in FOL event representation

  40. Selectional Restrictions • Semantic type constraint on arguments • I want to eat someplace close to UW • E.g. THEME of eating should be edible • Associated with senses • Vary in specificity: • Imagine: AGENT: human/sentient; THEME: any • Representation: • Add as predicate in FOL event representation • Overkill computationally; requires large commonsense KB

  41. Selectional Restrictions • Semantic type constraint on arguments • I want to eat someplace close to UW • E.g. THEME of eating should be edible • Associated with senses • Vary in specificity: • Imagine: AGENT: human/sentient; THEME: any • Representation: • Add as predicate in FOL event representation • Overkill computationally; requires large commonsense KB • Associate with WordNetsynset (and hyponyms)

  42. Primitive Decompositions • Jackendoff(1990), Dorr(1999), McCawley (1968) • Word meaning constructed from primitives • Fixed small set of basic primitives • E.g. cause, go, become, • kill=cause X to become Y

  43. Primitive Decompositions • Jackendoff(1990), Dorr(1999), McCawley (1968) • Word meaning constructed from primitives • Fixed small set of basic primitives • E.g. cause, go, become, • kill=cause X to become Y • Augment with open-ended “manner” • Y = not alive • E.g. walk vsrun

  44. Primitive Decompositions • Jackendoff(1990), Dorr(1999), McCawley (1968) • Word meaning constructed from primitives • Fixed small set of basic primitives • E.g. cause, go, become, • kill=cause X to become Y • Augment with open-ended “manner” • Y = not alive • E.g. walk vs run • Fixed primitives/Infinite descriptors

  45. Word Sense Disambiguation • SelectionalRestriction-based approaches • Limitations • Robust Approaches • Supervised Learning Approaches • Naïve Bayes • Dictionary-based Approaches • Bootstrapping Approaches • One sense per discourse/collocation • Unsupervised Approaches • Schutze’s word space • Resource-based Approaches • Dictionary parsing, WordNet Distance • Why they work • Why they don’t

  46. Word Sense Disambiguation • Application of lexical semantics • Goal: Given a word in context, identify the appropriate sense • E.g. plants and animals in the rainforest • Crucial for real syntactic & semantic analysis

  47. Word Sense Disambiguation • Application of lexical semantics • Goal: Given a word in context, identify the appropriate sense • E.g. plants and animals in the rainforest • Crucial for real syntactic & semantic analysis • Correct sense can determine • .

  48. Word Sense Disambiguation • Application of lexical semantics • Goal: Given a word in context, identify the appropriate sense • E.g. plants and animals in the rainforest • Crucial for real syntactic & semantic analysis • Correct sense can determine • Available syntactic structure • Available thematic roles, correct meaning,..

  49. Selectional Restriction Approaches • Integrate sense selection in parsing and semantic analysis – e.g. with lambda calculus • Concept: Predicate selects sense • Washing dishes vs stir-frying dishes

More Related